1
0
mirror of https://github.com/postgres/postgres.git synced 2025-07-30 11:03:19 +03:00

Add function to import operating system collations

Move this logic out of initdb into a user-callable function.  This
simplifies the code and makes it possible to update the standard
collations later on if additional operating system collations appear.

Reviewed-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Euler Taveira <euler@timbira.com.br>
This commit is contained in:
Peter Eisentraut
2017-01-18 12:00:00 -05:00
parent 193a7d791e
commit aa17c06fb5
8 changed files with 229 additions and 172 deletions

View File

@ -41,7 +41,8 @@ Oid
CollationCreate(const char *collname, Oid collnamespace,
Oid collowner,
int32 collencoding,
const char *collcollate, const char *collctype)
const char *collcollate, const char *collctype,
bool if_not_exists)
{
Relation rel;
TupleDesc tupDesc;
@ -72,10 +73,21 @@ CollationCreate(const char *collname, Oid collnamespace,
PointerGetDatum(collname),
Int32GetDatum(collencoding),
ObjectIdGetDatum(collnamespace)))
ereport(ERROR,
{
if (if_not_exists)
{
ereport(NOTICE,
(errcode(ERRCODE_DUPLICATE_OBJECT),
errmsg("collation \"%s\" for encoding \"%s\" already exists",
errmsg("collation \"%s\" for encoding \"%s\" already exists, skipping",
collname, pg_encoding_to_char(collencoding))));
return InvalidOid;
}
else
ereport(ERROR,
(errcode(ERRCODE_DUPLICATE_OBJECT),
errmsg("collation \"%s\" for encoding \"%s\" already exists",
collname, pg_encoding_to_char(collencoding))));
}
/*
* Also forbid matching an any-encoding entry. This test of course is not
@ -86,10 +98,21 @@ CollationCreate(const char *collname, Oid collnamespace,
PointerGetDatum(collname),
Int32GetDatum(-1),
ObjectIdGetDatum(collnamespace)))
ereport(ERROR,
{
if (if_not_exists)
{
ereport(NOTICE,
(errcode(ERRCODE_DUPLICATE_OBJECT),
errmsg("collation \"%s\" already exists, skipping",
collname)));
return InvalidOid;
}
else
ereport(ERROR,
(errcode(ERRCODE_DUPLICATE_OBJECT),
errmsg("collation \"%s\" already exists",
collname)));
}
/* open pg_collation */
rel = heap_open(CollationRelationId, RowExclusiveLock);

View File

@ -136,7 +136,11 @@ DefineCollation(ParseState *pstate, List *names, List *parameters)
GetUserId(),
GetDatabaseEncoding(),
collcollate,
collctype);
collctype,
false);
if (!OidIsValid(newoid))
return InvalidObjectAddress;
ObjectAddressSet(address, CollationRelationId, newoid);
@ -177,3 +181,151 @@ IsThereCollationInNamespace(const char *collname, Oid nspOid)
errmsg("collation \"%s\" already exists in schema \"%s\"",
collname, get_namespace_name(nspOid))));
}
/*
* "Normalize" a locale name, stripping off encoding tags such as
* ".utf8" (e.g., "en_US.utf8" -> "en_US", but "br_FR.iso885915@euro"
* -> "br_FR@euro"). Return true if a new, different name was
* generated.
*/
pg_attribute_unused()
static bool
normalize_locale_name(char *new, const char *old)
{
char *n = new;
const char *o = old;
bool changed = false;
while (*o)
{
if (*o == '.')
{
/* skip over encoding tag such as ".utf8" or ".UTF-8" */
o++;
while ((*o >= 'A' && *o <= 'Z')
|| (*o >= 'a' && *o <= 'z')
|| (*o >= '0' && *o <= '9')
|| (*o == '-'))
o++;
changed = true;
}
else
*n++ = *o++;
}
*n = '\0';
return changed;
}
Datum
pg_import_system_collations(PG_FUNCTION_ARGS)
{
#if defined(HAVE_LOCALE_T) && !defined(WIN32)
bool if_not_exists = PG_GETARG_BOOL(0);
Oid nspid = PG_GETARG_OID(1);
FILE *locale_a_handle;
char localebuf[NAMEDATALEN]; /* we assume ASCII so this is fine */
int count = 0;
#endif
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to import system collations"))));
#if defined(HAVE_LOCALE_T) && !defined(WIN32)
locale_a_handle = OpenPipeStream("locale -a", "r");
if (locale_a_handle == NULL)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not execute command \"%s\": %m",
"locale -a")));
while (fgets(localebuf, sizeof(localebuf), locale_a_handle))
{
int i;
size_t len;
int enc;
bool skip;
char alias[NAMEDATALEN];
len = strlen(localebuf);
if (len == 0 || localebuf[len - 1] != '\n')
{
elog(DEBUG1, "locale name too long, skipped: \"%s\"", localebuf);
continue;
}
localebuf[len - 1] = '\0';
/*
* Some systems have locale names that don't consist entirely of ASCII
* letters (such as "bokm&aring;l" or "fran&ccedil;ais"). This is
* pretty silly, since we need the locale itself to interpret the
* non-ASCII characters. We can't do much with those, so we filter
* them out.
*/
skip = false;
for (i = 0; i < len; i++)
{
if (IS_HIGHBIT_SET(localebuf[i]))
{
skip = true;
break;
}
}
if (skip)
{
elog(DEBUG1, "locale name has non-ASCII characters, skipped: \"%s\"", localebuf);
continue;
}
enc = pg_get_encoding_from_locale(localebuf, false);
if (enc < 0)
{
/* error message printed by pg_get_encoding_from_locale() */
continue;
}
if (!PG_VALID_BE_ENCODING(enc))
continue; /* ignore locales for client-only encodings */
if (enc == PG_SQL_ASCII)
continue; /* C/POSIX are already in the catalog */
count++;
CollationCreate(localebuf, nspid, GetUserId(), enc,
localebuf, localebuf, if_not_exists);
CommandCounterIncrement();
/*
* Generate aliases such as "en_US" in addition to "en_US.utf8" for
* ease of use. Note that collation names are unique per encoding
* only, so this doesn't clash with "en_US" for LATIN1, say.
*
* This always runs in "if not exists" mode, to skip aliases that
* conflict with an existing locale name for the same encoding. For
* example, "br_FR.iso88591" is normalized to "br_FR", both for
* encoding LATIN1. But the unnormalized locale "br_FR" already
* exists for LATIN1.
*/
if (normalize_locale_name(alias, localebuf))
{
CollationCreate(alias, nspid, GetUserId(), enc,
localebuf, localebuf, true);
CommandCounterIncrement();
}
}
ClosePipeStream(locale_a_handle);
if (count == 0)
ereport(ERROR,
(errmsg("no usable system locales were found")));
#endif /* not HAVE_LOCALE_T && not WIN32 */
PG_RETURN_VOID();
}

View File

@ -1608,178 +1608,16 @@ setup_description(FILE *cmdfd)
PG_CMD_PUTS("DROP TABLE tmp_pg_shdescription;\n\n");
}
#ifdef HAVE_LOCALE_T
/*
* "Normalize" a locale name, stripping off encoding tags such as
* ".utf8" (e.g., "en_US.utf8" -> "en_US", but "br_FR.iso885915@euro"
* -> "br_FR@euro"). Return true if a new, different name was
* generated.
*/
static bool
normalize_locale_name(char *new, const char *old)
{
char *n = new;
const char *o = old;
bool changed = false;
while (*o)
{
if (*o == '.')
{
/* skip over encoding tag such as ".utf8" or ".UTF-8" */
o++;
while ((*o >= 'A' && *o <= 'Z')
|| (*o >= 'a' && *o <= 'z')
|| (*o >= '0' && *o <= '9')
|| (*o == '-'))
o++;
changed = true;
}
else
*n++ = *o++;
}
*n = '\0';
return changed;
}
#endif /* HAVE_LOCALE_T */
/*
* populate pg_collation
*/
static void
setup_collation(FILE *cmdfd)
{
#if defined(HAVE_LOCALE_T) && !defined(WIN32)
int i;
FILE *locale_a_handle;
char localebuf[NAMEDATALEN]; /* we assume ASCII so this is fine */
int count = 0;
locale_a_handle = popen_check("locale -a", "r");
if (!locale_a_handle)
return; /* complaint already printed */
PG_CMD_PUTS("CREATE TEMP TABLE tmp_pg_collation ( "
" collname name, "
" locale name, "
" encoding int) WITHOUT OIDS;\n\n");
while (fgets(localebuf, sizeof(localebuf), locale_a_handle))
{
size_t len;
int enc;
bool skip;
char *quoted_locale;
char alias[NAMEDATALEN];
len = strlen(localebuf);
if (len == 0 || localebuf[len - 1] != '\n')
{
if (debug)
fprintf(stderr, _("%s: locale name too long, skipped: \"%s\"\n"),
progname, localebuf);
continue;
}
localebuf[len - 1] = '\0';
/*
* Some systems have locale names that don't consist entirely of ASCII
* letters (such as "bokm&aring;l" or "fran&ccedil;ais"). This is
* pretty silly, since we need the locale itself to interpret the
* non-ASCII characters. We can't do much with those, so we filter
* them out.
*/
skip = false;
for (i = 0; i < len; i++)
{
if (IS_HIGHBIT_SET(localebuf[i]))
{
skip = true;
break;
}
}
if (skip)
{
if (debug)
fprintf(stderr, _("%s: locale name has non-ASCII characters, skipped: \"%s\"\n"),
progname, localebuf);
continue;
}
enc = pg_get_encoding_from_locale(localebuf, debug);
if (enc < 0)
{
/* error message printed by pg_get_encoding_from_locale() */
continue;
}
if (!PG_VALID_BE_ENCODING(enc))
continue; /* ignore locales for client-only encodings */
if (enc == PG_SQL_ASCII)
continue; /* C/POSIX are already in the catalog */
count++;
quoted_locale = escape_quotes(localebuf);
PG_CMD_PRINTF3("INSERT INTO tmp_pg_collation VALUES (E'%s', E'%s', %d);\n\n",
quoted_locale, quoted_locale, enc);
/*
* Generate aliases such as "en_US" in addition to "en_US.utf8" for
* ease of use. Note that collation names are unique per encoding
* only, so this doesn't clash with "en_US" for LATIN1, say.
*/
if (normalize_locale_name(alias, localebuf))
{
char *quoted_alias = escape_quotes(alias);
PG_CMD_PRINTF3("INSERT INTO tmp_pg_collation VALUES (E'%s', E'%s', %d);\n\n",
quoted_alias, quoted_locale, enc);
free(quoted_alias);
}
free(quoted_locale);
}
PG_CMD_PUTS("SELECT pg_import_system_collations(if_not_exists => false, schema => 'pg_catalog');\n\n");
/* Add an SQL-standard name */
PG_CMD_PRINTF1("INSERT INTO tmp_pg_collation VALUES ('ucs_basic', 'C', %d);\n\n", PG_UTF8);
/*
* When copying collations to the final location, eliminate aliases that
* conflict with an existing locale name for the same encoding. For
* example, "br_FR.iso88591" is normalized to "br_FR", both for encoding
* LATIN1. But the unnormalized locale "br_FR" already exists for LATIN1.
* Prefer the alias that matches the OS locale name, else the first locale
* name by sort order (arbitrary choice to be deterministic).
*
* Also, eliminate any aliases that conflict with pg_collation's
* hard-wired entries for "C" etc.
*/
PG_CMD_PUTS("INSERT INTO pg_collation (collname, collnamespace, collowner, collencoding, collcollate, collctype) "
" SELECT DISTINCT ON (collname, encoding)"
" collname, "
" (SELECT oid FROM pg_namespace WHERE nspname = 'pg_catalog') AS collnamespace, "
" (SELECT relowner FROM pg_class WHERE relname = 'pg_collation') AS collowner, "
" encoding, locale, locale "
" FROM tmp_pg_collation"
" WHERE NOT EXISTS (SELECT 1 FROM pg_collation WHERE collname = tmp_pg_collation.collname)"
" ORDER BY collname, encoding, (collname = locale) DESC, locale;\n\n");
/*
* Even though the table is temp, drop it explicitly so it doesn't get
* copied into template0/postgres databases.
*/
PG_CMD_PUTS("DROP TABLE tmp_pg_collation;\n\n");
pclose(locale_a_handle);
if (count == 0 && !debug)
{
printf(_("No usable system locales were found.\n"));
printf(_("Use the option \"--debug\" to see details.\n"));
}
#endif /* not HAVE_LOCALE_T && not WIN32 */
PG_CMD_PRINTF2("INSERT INTO pg_collation (collname, collnamespace, collowner, collencoding, collcollate, collctype) VALUES ('ucs_basic', 'pg_catalog'::regnamespace, '%s'::regrole, %d, 'C', 'C');\n\n", escape_quotes(username), PG_UTF8);
}
/*

View File

@ -53,6 +53,6 @@
*/
/* yyyymmddN */
#define CATALOG_VERSION_NO 201701172
#define CATALOG_VERSION_NO 201701181
#endif

View File

@ -17,7 +17,8 @@
extern Oid CollationCreate(const char *collname, Oid collnamespace,
Oid collowner,
int32 collencoding,
const char *collcollate, const char *collctype);
const char *collcollate, const char *collctype,
bool if_not_exists);
extern void RemoveCollationById(Oid collationOid);
#endif /* PG_COLLATION_FN_H */

View File

@ -5349,6 +5349,9 @@ DESCR("pg_controldata recovery state information as a function");
DATA(insert OID = 3444 ( pg_control_init PGNSP PGUID 12 1 0 0 0 f f f f t f v s 0 0 2249 "" "{23,23,23,23,23,23,23,23,23,16,16,16,23}" "{o,o,o,o,o,o,o,o,o,o,o,o,o}" "{max_data_alignment,database_block_size,blocks_per_segment,wal_block_size,bytes_per_wal_segment,max_identifier_length,max_index_columns,max_toast_chunk_size,large_object_chunk_size,bigint_timestamps,float4_pass_by_value,float8_pass_by_value,data_page_checksum_version}" _null_ _null_ pg_control_init _null_ _null_ _null_ ));
DESCR("pg_controldata init state information as a function");
DATA(insert OID = 3445 ( pg_import_system_collations PGNSP PGUID 12 100 0 0 0 f f f f t f v r 2 0 2278 "16 4089" _null_ _null_ "{if_not_exists,schema}" _null_ _null_ pg_import_system_collations _null_ _null_ _null_ ));
DESCR("import collations from operating system");
/*
* Symbolic values for provolatile column: these indicate whether the result
* of a function is dependent *only* on the values of its explicit arguments,